[00:09:27] *** Joins: tomzawadzki (~tomzawadz@134.134.139.83) [00:13:00] *** Quits: Shuhei (caf6fc61@gateway/web/freenode/ip.202.246.252.97) (Ping timeout: 252 seconds) [01:29:47] *** Joins: wzh (~wzh@114.255.44.139) [01:31:06] *** Quits: tomzawadzki (~tomzawadz@134.134.139.83) (Read error: Connection reset by peer) [02:21:19] *** Joins: vmysak (~vmysak@134.134.139.75) [03:27:11] *** Quits: wzh (~wzh@114.255.44.139) (Quit: WeeChat 1.9.1) [05:02:30] *** Quits: vmysak (~vmysak@134.134.139.75) (Ping timeout: 246 seconds) [05:54:59] *** Joins: tomzawadzki (tomzawadzk@nat/intel/x-upzyuwbcmaheiymw) [07:59:51] Hello everyone. I have a question. Why do we call spdk_io_device_register() in some bdevs (e.g. in null or aio) and do not do so in others (e.g malloc) ? [08:30:40] *** Joins: vmysak (~vmysak@134.134.139.73) [08:34:03] you only need to call spdk_io_device_register if you need to create channels (per-thread contexts) [08:34:34] malloc is just doing memcpy essentially, so it doesn't need any per-thread context [08:54:10] thanks [09:34:40] *** Quits: tomzawadzki (tomzawadzk@nat/intel/x-upzyuwbcmaheiymw) (Ping timeout: 246 seconds) [09:35:04] *** Quits: vmysak (~vmysak@134.134.139.73) (Ping timeout: 240 seconds) [09:58:49] *** Joins: travis-ci (~travis-ci@ec2-54-224-21-99.compute-1.amazonaws.com) [09:58:50] (spdk/master) bdev/gpt: vbdev_gpt_examine: no need to read gpt if bdev is too small (wuzhouhui) [09:58:50] Diff URL: https://github.com/spdk/spdk/compare/b98bbb18c65e...e133dcd85040 [09:58:50] *** Parts: travis-ci (~travis-ci@ec2-54-224-21-99.compute-1.amazonaws.com) () [10:19:56] *** Joins: bluebird (~bluebird@p5DE955DE.dip0.t-ipconnect.de) [10:30:02] *** Joins: darsto (~darsto@89-78-174-111.dynamic.chello.pl) [12:06:01] *** Joins: vmysak (vmysak@nat/intel/x-isxseeozyqotvzrh) [12:36:18] *** Quits: vmysak (vmysak@nat/intel/x-isxseeozyqotvzrh) (Ping timeout: 245 seconds) [13:05:12] https://ci.spdk.io/spdk/builds/review/e0c71bbab92bbcfb53c2b0305a356fcab7776143.1536262880/fedora-02/build.log [13:06:09] https://ci.spdk.io/spdk/builds/review/e0c71bbab92bbcfb53c2b0305a356fcab7776143.1536262880/ubuntu17.10/build.log [13:29:39] *** Quits: bluebird (~bluebird@p5DE955DE.dip0.t-ipconnect.de) (Quit: Leaving) [13:47:11] *** Joins: travis-ci (~travis-ci@ec2-54-224-21-99.compute-1.amazonaws.com) [13:47:12] (spdk/master) test: re-enable bdevperf write_zeroes test (Jim Harris) [13:47:12] Diff URL: https://github.com/spdk/spdk/compare/e133dcd85040...71abba4eaf81 [13:47:12] *** Parts: travis-ci (~travis-ci@ec2-54-224-21-99.compute-1.amazonaws.com) () [13:53:52] sethhowe: Saw your comment, https://github.com/spdk/spdk/issues/426#issuecomment-419235366 regarding shared lib build, but was confused by the posted gerrithub link therein (nvme_rdma: Check mr size...) [13:57:29] lhodev: Thanks! Just fixed it to point to the right one. [14:01:52] *** Joins: travis-ci (~travis-ci@ec2-54-224-21-99.compute-1.amazonaws.com) [14:01:53] (spdk/master) examples/ioat/verify: improve get_next_chan() error message (Jim Harris) [14:01:53] Diff URL: https://github.com/spdk/spdk/compare/71abba4eaf81...c59205543c94 [14:01:53] *** Parts: travis-ci (~travis-ci@ec2-54-224-21-99.compute-1.amazonaws.com) () [14:55:32] OK, about to try and get fio working w/crypto.... wish me something (luck, pain, etc) [15:26:55] so this is unexpected... I got fio to work with straight nvme. when I add in crypto on top I get a DPDK function failure that I've never seen before and the same built code works fine with our other tools. It's a failure in rte_zmalloc_socket() to get memory when the crypto device is setting up its Q pairs. [15:27:05] Not sure how to start with LD_PRELOAD and gdb... any tips? [15:34:11] hmmm, got passed that. when I call the SW device crypto init API I wasn't specifying a socket and for our tools that was fine. Starting with FIO I have to specify the socket or stuff doesn't work. Seems a little strange, hardcoding 0 for now. Next issue is with claiming, does shit just work totally differently w/the fio plugin or what? [15:34:53] duh, never mind [15:35:45] had to change my spdk .conf file as well as the fio conf file to point to the crypto vbdev. can run basic r/w/verify now, sstill a little confused about the socket thing so will look further into that.... [16:45:57] *** Joins: travis-ci (~travis-ci@ec2-54-82-100-208.compute-1.amazonaws.com) [16:45:58] (spdk/master) bdev: set iovs on correct bdev_io in spdk_bdev_io_put_buf (Jim Harris) [16:45:58] Diff URL: https://github.com/spdk/spdk/compare/f118de60af03...de4d961996d7 [16:45:58] *** Parts: travis-ci (~travis-ci@ec2-54-82-100-208.compute-1.amazonaws.com) () [17:03:15] *** Joins: travis-ci (~travis-ci@ec2-54-82-100-208.compute-1.amazonaws.com) [17:03:16] (spdk/master) build: add quiet output for recently added shared lib build (Jim Harris) [17:03:16] Diff URL: https://github.com/spdk/spdk/compare/de4d961996d7...cc5fedc6ebb8 [17:03:16] *** Parts: travis-ci (~travis-ci@ec2-54-82-100-208.compute-1.amazonaws.com) () [17:40:17] *** Joins: travis-ci (~travis-ci@ec2-54-224-21-99.compute-1.amazonaws.com) [17:40:18] (spdk/master) nvme: move controller setting into nvme_ctrlr_identify() (Changpeng Liu) [17:40:18] Diff URL: https://github.com/spdk/spdk/compare/cc5fedc6ebb8...cab35b6bc1e3 [17:40:18] *** Parts: travis-ci (~travis-ci@ec2-54-224-21-99.compute-1.amazonaws.com) () [18:03:35] *** Joins: wzh (~wzh@114.255.44.139) [18:06:02] bwalker: are you still online? I have leave a comment in patch https://review.gerrithub.io/c/spdk/spdk/+/423369, I don't know whether you have got the notification email. [18:33:09] jimharris, FYI latest crypto asserted in CI on incoming IO being larger than it should have been based on the block limits I put in. Have to head out to dinner with fam but will get to the bottom of it in the morn, it didn't fail for me like that locally... [19:31:04] *** Quits: wzh (~wzh@114.255.44.139) (Quit: WeeChat 1.9.1) [19:33:48] *** Joins: travis-ci (~travis-ci@ec2-54-224-21-99.compute-1.amazonaws.com) [19:33:49] (spdk/master) nvme: return errno for nvme_ctrlr_set_supported_log_pages() (Changpeng Liu) [19:33:49] Diff URL: https://github.com/spdk/spdk/compare/cab35b6bc1e3...9ad05b31425e [19:33:49] *** Parts: travis-ci (~travis-ci@ec2-54-224-21-99.compute-1.amazonaws.com) ()